SHARP: hyperfast and accurate processing of single-cell RNA-seq data via ensemble random projection.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) generates massive datasets.
• Dimension reduction is crucial for scRNA-seq data analysis but can cause distortion.
• Existing methods struggle with scalability and accuracy for large scRNA-seq datasets.

Purpose of the Study:

• To develop a scalable and accurate algorithm for processing large-scale scRNA-seq data.
• To address the computational challenges of dimension reduction in scRNA-seq analysis.
• To provide an efficient tool for clustering millions of single cells.

Main Methods:

• SHARP (ensemble random projection-based algorithm)
• Benchmarking on 17 public scRNA-seq datasets
• Scalability testing up to 10 million cells

Main Results:

• SHARP demonstrates superior speed and accuracy compared to existing methods.
• For datasets >40,000 cells, SHARP is faster while maintaining high clustering accuracy.
• SHARP is the first R-based tool scalable to clustering 10 million scRNA-seq cells.

Conclusions:

• SHARP offers an effective solution for large-scale scRNA-seq data processing.
• The algorithm provides a scalable and robust approach to dimension reduction and clustering.
• SHARP enhances the analysis of complex single-cell genomics studies.